MichaelGrupp / evo

Python package for the evaluation of odometry and SLAM
https://michaelgrupp.github.io/evo/
GNU General Public License v3.0
3.43k stars 746 forks source link

RPE values is wrong #625

Closed BamsaAhmed closed 8 months ago

BamsaAhmed commented 8 months ago

I used evo for find APE and RPE for my SLAM, the ape give correct value and curve, while the rpe curve is not correct for the same data? Both curves is shown below: image rmse13r1

could anyone tell me why?

BamsaAhmed commented 8 months ago

note the color not changed along the path for rpe and the values are: for RPE: RPE w.r.t. translation part (m) for delta = 1 (frames) using consecutive pairs (with SE(3) Umeyama alignment)

   max  1.005466
  mean  0.044835
median  0.034352
   min  0.000965
  rmse  0.060864
   sse  25.338594
   std  0.041162

while for APE: APE w.r.t. translation part (m) (not aligned)

   max  360.950139
  mean  312.292985
median  311.861015
   min  257.837880
  rmse  313.705472
   sse  673230491.551141
   std  29.735741
LeeBY68 commented 8 months ago

seem your rpe visualization did not aligned.

MichaelGrupp commented 8 months ago

Without specifying why you think it's "wrong", we can't help you. With the current information given, it looks like you might just be confused by the missing alignment as mentioned by @LeeBY68 . Feel free to re-open if you have specific questions.